A series of experiments is proposed to investigate the basic speech perception capacities of infants and the role of these capacities in acquiring a native language. During the course of development, speech-processing capacities become adapted to the specific organization of native language sound patterns to provide fast and efficient recognition of words in fluent speech. Accordingly, one aim of the proposed research is to develop a clearer picture of the speech information that infants encode and remember. A fuller account of infants' memory capacities for speech is required to understand processes underlying word learning and the development of a lexicon. A number of the proposed studies examine the nature and extent of information that infants retain about speech and the implications this has for the growth and organization of a lexicon. The investigator has been exploring infants' use of their growing knowledge of native language sounds patterns to segment words from fluent speech. Thus, a second aim is to identify potential cues to word boundaries used by infants in processing fluent speech and to investigate how and when infants integrate the various cues that are available to them. A clearer picture of the nature of the words that infants detect and encode from fluent speech will be useful in understanding how they begin to discover the syntactic organization of native language utterances. For example, sensitivity to the occurrence of positioning of certain grammatical morphemes in utterances could provide the learner with clues about syntactic organization. Previous research from this laboratory has shown that infants are sensitive to the prosodic marking of utterances. Thus, a third major aim of the proposed research how this sensitivity to prosody might be used with other information in the speech signal, such as the occurrence of grammatical morphemes to aid learners in the discovery of the syntactic organization of their native language.